Path Planning Based on YOLOX and Improved Dynamic Window Approach

In recent years, mobile robot technology has developed rapidly. Path planning is an important part of mobile robot design. Based on this technology, autonomous decision-making navigation of robots can be realized, so as to reach the predetermined target position smoothly and efficiently. The conventional path planning algorithm is generally a static trajectory adjustment algorithm. However, the surrounding environment is complex and changeable during the actual movement of the robot. It is difficult to ensure real-time performance for the static algorithm when encountering unknown situations. Therefore, the Dynamic Window Approach is studied and improved for the unknown environment that the robot may face, and YOLOX is present to detect obstacles. By setting different safe distances for different obstacles, a more intelligent robot dynamic trajectory adjustment is achieved. Finally, the simulation experiment is carried out in MATLAB. By comparison, the improved Dynamic Window Approach obtains a more secure and time-saving moving trajectory, which proves the effectiveness of the algorithm.